drm/i915: Tighten reset_counter for reset status

In the reset_counter, we use two bits to track a GPU hang and reset. The
low bit is a "reset-in-progress" flag that we set to signal when we need
to break waiters in order for the recovery task to grab the mutex. As
soon as the recovery task has the mutex, we can clear that flag (which
we do by incrementing the reset_counter thereby incrementing the gobal
reset epoch). By clearing that flag when the recovery task holds the
struct_mutex, we can forgo a second flag that simply tells GEM to ignore
the "reset-in-progress" flag.

The second flag we store in the reset_counter is whether the
reset failed and we consider the GPU terminally wedged. Whilst this flag
is set, all access to the GPU (at least through GEM rather than direct mmio
access) is verboten.

PS: Fun is in store, as in the future we want to move from a global
reset epoch to a per-engine reset engine with request recovery.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1460565315-7748-6-git-send-email-chris@chris-wilson.co.uk

1 files changed, 23 insertions, 16 deletions

diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 1dca3442c545..633d0ddace6a 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -880,23 +880,32 @@ int i915_resume_switcheroo(struct drm_device *dev)
 int i915_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	bool simulated;
+	struct i915_gpu_error *error = &dev_priv->gpu_error;
+	unsigned reset_counter;
 	int ret;
 
 	intel_reset_gt_powersave(dev);
 
 	mutex_lock(&dev->struct_mutex);
 
-	i915_gem_reset(dev);
+	/* Clear any previous failed attempts at recovery. Time to try again. */
+	atomic_andnot(I915_WEDGED, &error->reset_counter);
 
-	simulated = dev_priv->gpu_error.stop_rings != 0;
+	/* Clear the reset-in-progress flag and increment the reset epoch. */
+	reset_counter = atomic_inc_return(&error->reset_counter);
+	if (WARN_ON(__i915_reset_in_progress(reset_counter))) {
+		ret = -EIO;
+		goto error;
+	}
+
+	i915_gem_reset(dev);
 
 	ret = intel_gpu_reset(dev, ALL_ENGINES);
 
 	/* Also reset the gpu hangman. */
-	if (simulated) {
+	if (error->stop_rings != 0) {
 		DRM_INFO("Simulated gpu hang, resetting stop_rings\n");
-		dev_priv->gpu_error.stop_rings = 0;
+		error->stop_rings = 0;
 		if (ret == -ENODEV) {
 			DRM_INFO("Reset not implemented, but ignoring "
 				 "error for simulated gpu hangs\n");
@@ -909,8 +918,7 @@ int i915_reset(struct drm_device *dev)
 
 	if (ret) {
 		DRM_ERROR("Failed to reset chip: %i\n", ret);
-		mutex_unlock(&dev->struct_mutex);
-		return ret;
+		goto error;
 	}
 
 	intel_overlay_reset(dev_priv);
@@ -929,20 +937,14 @@ int i915_reset(struct drm_device *dev)
 	 * was running at the time of the reset (i.e. we weren't VT
 	 * switched away).
 	 */
-
-	/* Used to prevent gem_check_wedged returning -EAGAIN during gpu reset */
-	dev_priv->gpu_error.reload_in_reset = true;
-
 	ret = i915_gem_init_hw(dev);
-
-	dev_priv->gpu_error.reload_in_reset = false;
-
-	mutex_unlock(&dev->struct_mutex);
 	if (ret) {
 		DRM_ERROR("Failed hw init on reset %d\n", ret);
-		return ret;
+		goto error;
 	}
 
+	mutex_unlock(&dev->struct_mutex);
+
 	/*
 	 * rps/rc6 re-init is necessary to restore state lost after the
 	 * reset and the re-install of gt irqs. Skip for ironlake per
@@ -953,6 +955,11 @@ int i915_reset(struct drm_device *dev)
 		intel_enable_gt_powersave(dev);
 
 	return 0;
+
+error:
+	atomic_or(I915_WEDGED, &error->reset_counter);
+	mutex_unlock(&dev->struct_mutex);
+	return ret;
 }
 
 static int i915_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)